1. Field of the Invention
The present invention relates to techniques for embedding digital watermark information mainly in document images.
2. Description of the Related Art
Recently, in the trend of networking and digitization of media for text data, image data, audio data, and the like, demand has arisen for prevention of unauthorized copying of digital data and for association of meta data with digital data. Still, however, documents and images are often distributed in the form of printed materials. Since both digital data and printed materials are used as described above, demand has also arisen for techniques for controlling a destination to which digital data is to be distributed in the form of a printed material or linking a printed material with digital data. Under such circumstances, interest has arisen for a method for embedding information by digitally watermarking a multi-level still image or document image.
Embedding of information by digital watermarking is a technique for modifying part of original data to embed information, in a human-unperceivable manner, in digital image data, audio data, or text data.
For example, as techniques for embedding and extracting a digital watermark in and from a multi-level image, various methods based on the redundancy of the intensity levels of pixels are generally known. Also, as a technique for digitally watermarking a document image that is a binary image, several digital watermarking methods based on characteristics specific to document images are known. For example, a method of embedding digital watermark information by changing the length of a space between words of English text (alphabetic text) according to a predetermined criterion is known. Such a method is disclosed, for example, in Japanese Patent Laid-Open No. 9-186603 (U.S. Pat. No. 5,861,619).
According to the digital watermarking method for documents, digital watermark information is represented by varying the lengths of spaces between words or between characters so that one-bit information (either “1” or “0”) will be assigned according to the values of the two space lengths.
In another method, digital watermark information is embedded in a character by rotating the character to change a tilt angle thereof. Such techniques are disclosed, for example, in Nakamura and Matsui, “Wa-bunsho heno shiirugazou niyoru denshisukashi” (which could be translated as “Digital Watermarking of Japanese Documents using Seal Images”), November 1997 (Transactions of IPSJ, Vol. 38-No. 11).
A document image in which digital watermark information is embedded according to the related art will be described with reference to FIG. 13. FIG. 13 is a schematic diagram showing an example of a document image in which digital watermark information has been embedded by rotating a character to change a tilt angle thereof according to the related art.
For example, if a character is rotated clockwise, “1” is embedded in the character (as in character (1) in FIG. 13), and if a character is rotated counterclockwise, “0” is embedded in the character (as in character (2) in FIG. 13). Digital watermark may be embedded in continuous characters, at an interval of several characters, or in characters at predetermined positions.
Referring to FIG. 13, since the character (1) is rotated clockwise and the character (2) is rotated counterclockwise, information representing “10” is embedded.
However, regarding embedding of digital watermark information in document images according to the related art, a method of embedding a digital watermark with consideration of the layout of characters in the document image has not been established. For example, when a digital watermark is embedded in a document image by the method according to the related art, a feature regarding each character or each character and an adjacent space is considered. When the layout of the document image is complex, for example, when illustrations, tables and figures, and text areas are mixed or when the document image is multicolumned, if digital watermark information is embedded simply in each of the areas, the digital watermark information becomes discontinuous among the areas. Thus, it is not allowed to embed digital watermark information continuously, which is inconvenient.